What makes for effective and meaningful parliamentary online public engagement?

In my eyes this project involves going back to basics. The aim is to emulate a natural face to face discussion, over the internet. Naturally when we are conversing, our brain will do all the hard work in analysing someone's words, actions and body language, and tone to decipher what they really mean from what comes out of their mouth. This will then influence the subsequent decisions we make. This becomes increasingly difficult through written text, and as such, existing systems in place for capturing and analysing this data are not adequate. Therefore, the role of this project is to create a system in which real insights can be gleaned directly from the words of the people we need to listen to most, the citezans. As a data scientist, understanding not only how to collect the data, but what it means on a grander scale or the big picture is crucial in completing the job. As a linguist, I understand the underlying elements of not just English, but all natural language, how to map that into predicate logic computations and rules, and the pragmatics that feature in language. I intend to combine the knowledge from these two disciplines to bring a novel perspective to this project, one that does not just focus on the politics, computing, or statistical theory, but always has the different areas in mind, ready to create a well-rounded and sustainable product. The project can be broken down by years; in the first year I aim to have a good grasp on the processes already in place, understanding the business problem at hand from the people who will be using the tool on a daily basis, and the opportunity to collect and analyse the existing data to complete quality and missingness checks. In the second year, I aim to create some preliminary exploratory tools to better capture the data and improve analysis. These will be tested and the results analysed and compared, and in the final year I will use those results to create an implement the completed tool for future use. As I predict the volume of data to be large and constant, I would look to use a big data system such as Apache Spark with Hadoop and Apache Streaming for real-time data analysis if necessary. This would of course depend on the existing system in place and how it integrates.